d 10 1 mapping flexibility and security performance in the ... · our more detailed policy analysis...

This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under Grant Agreement no. 613256.

D 10.1 - Mapping Flexibility and Security Performance in the Face of the Crisis Raul Eamets and Katrin Humal University of Tartu Miroslav Beblavý and Ilaria Maselli Centre for European Policy Studies Kariappa Bheemaiah and Mark Smith Grenoble Ecole de Management Mairead Finn and Janine Leschke Copenhagen Business School WP10 - Flexicurity Version – FINAL 29 May 2015

2 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith STYLE Working Papers are peer-reviewed outputs from the www.style-research.eu project. The series is edited by the project coordinator Professor Jacqueline O’Reilly. These working papers are intended to meet the European Commission’s expected impact from the project:

i) to ‘advance the knowledge base that underpins the formulation and implementation of relevant policies in Europe with the aim of enhancing the employment of young people and their transition to economic and social independence’, and

ii) to engage with ‘relevant communities, stakeholders and practitioners in the research with a view to supporting employment policies in Europe.’ Contributions to a dialogue about these results can be made through the project website www.style-research.eu, or by following us on twitter @STYLEEU.

To cite this report: Eamets, R., Beblavý, M., Bheemaiah, K., Finn, M., Humal, K., Leschke, J., Maselli, I. and Smith, M (2015) Mapping Flexibility and Security Performance in the Face of the Crisis, STYLE Working Papers, WP10.1. CROME, University of Brighton, Brighton. http://www.style-research.eu/publications/working-papers/ © Copyright is held by the authors About the authors Raul Eamets, University of Tartu www.style-research.eu/team/raul-eamets/ (Sections 1, 2, 5) Miroslav Beblavý, Centre for European Policy Studies www.style-research.eu/team/miroslav-beblavy/ (Section 4) Kariappa Bheemaiah, Grenoble Ecole de Management (Section 2.2) Mairead Finn, Copenhagen Business School (Section 3) Katrin Humal, University of Tartu (Sections 1, 2.1, 5) Janine Leschke, Copenhagen Business School www.style-research.eu/team/janine-leschke/ (Section 3) Ilaria Maselli, Centre for European Policy Studies www.style-research.eu/team/ilaria-maselli/ (Section 4) Mark Smith, Grenoble Ecole de Management www.style-research.eu/team/mark-smith/ (Sections 1, 2, 5) Acknowledgements The research leading to these results has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 613256. The authors would like to thank Jacqueline O’Reilly and the coordination team at the University of Brighton and Paola Villa at the University of Trento for their guidance in coordinating this Work Package. We would particularly like to thank the European Commission’s DG for Employment, Social Affairs and Inclusion, DG for Research and Innovation, Maria Jepsen from the European Trade Union Institute and Brendan Burchell at the University of Cambridge for their helpful reviews of earlier drafts.

D 10.1 - Mapping Flexibility and Security Performance in the Face of the Crisis 3

Executive Summary The term Flexicurity was a dominant theme at the European and national level in the first decade of this century and although its use has declined in recent years, the balance of flexibility and security for labour market participants remains a central factor in determining labour market outcomes –particularly for young people. The concept has, however, received a lot of criticism, often related to a lack of clarity in definition. On the one hand, the definitional ambiguity helps explains why the concept was picked up so easily at the policy level across a wide variety of stakeholders and contexts. But on the other hand, the ambiguity also explains how policies resulting in an overemphasis on (external) flexibility and employability, with little emphasis on job and income security, have developed. Thus the balance of flexibility and security remains a key dimension in understanding the plight of young people entering the labour market and the economic crisis has only served to further expose the uneven security afforded to different labour market groups. It is with this broader concept of “Flexicurity” – as the balance between flexibility and security on the labour market – that we approach the theme in this report. Youth tends to accumulate negative flexibility outcomes in that they have more limited contractual security, a greater risk of working on non-standard contracts and may lose their jobs more quickly than the comparable adult population. At the same time young people also have less job and income security due to their lower seniority and more limited employment histories. Furthermore in most European countries workers on non-standard contracts have more limited access to unemployment benefits than workers on standard employment contracts, which can exacerbate the position of vulnerable labour market groups often disproportionally engaged on such contracts – young people, women and people with lower education levels. The crisis further emphasised the risks of these negative outcomes. In order to explore this balance between flexibility and security for young people we undertake a number of analyses in this report. Firstly, we carry out a detailed mapping of flexibility-security indicators for European Union member states in order to identify clusters of policy making and flexibility-security outcomes, including key trends during the crisis period. Secondly, we explore the evolution of the income security measure on the labour market for young people and their links with internal and external flexibility outcomes. This exercise highlights some of the tensions and contradictions that exist in policy making while also charting the direction of travel for policy aimed at youth labour markets in recent years. Lastly, we present a detailed analysis of the performance of European labour markets that compares unemployment outcomes for younger and older workers and highlights the weaknesses of some conventional measures when applied to the youth labour market. The mapping exercise, using national-level data from the OECD and Eurostat, shows that in terms of institutional settings of flexibility and security East European countries consistently group together, as do Nordic countries with the Netherlands and Germany. These results are generally in line with earlier attempts to cluster flexibility-security regimes (EC 2006; Philips & Eamets, 2007; Auer, 2010; Chung, 2012). A challenge for such analyses remains the data limitations. Comparison of selected national-level outcome indicators shows that country groups with similar institutional settings do not necessarily have similar labour market and/or social outcomes for young people. This supports earlier findings that institutional and outcome-type measures of flexibility and security might not be correlated and should be examined separately (Chung, 2012). These results suggest that there are a range of forces shaping outcomes on the youth labour market and within-cluster, and indeed within-country, variations need to be taken into account in explaining the variety of outcomes – further underlining for policy makers and researchers alike that youth are far from a homogenous group. Overall young people tend to have worse flexibility-security outcomes, especially after the initial effects of the crisis. This is in line with previous literature indicating that vulnerable groups on the labour market, such as youth, the elderly, women, the long-term unemployed and temporary employees, do not experience the same wins that regular employees might gain from flexibility-security policies (for example Maselli, 2010; Leschke, 2012).

4 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith Our more detailed policy analysis of security measure, making use of Eurostat LFS special extracts and other sources, shows that it is important not to limit the analytical lens to youth between 15 and 24 and equally not to simply merge the younger and older youth groups (25-29). We show that older youth are better off than younger ones in terms of external – but not internal – numerical flexibility. We do, however, lack detailed, age-specific information on some internal security measures such as compensation for short-time working. Older youth are also better off than the youngest with regard to income security yet both groups of youth do worse than adults on all three dimensions.

These policy analyses emphasise the complexity of unemployment benefit schemes across Europe– particularly in terms of coverage, generosity, and the availability of secondary schemes, combined with the impact of frequent and not necessarily coherent adjustments during the first part economic crisis Thus, comparative analysis of various aspects of access to benefits is difficult. Attempts to create “simple” indices on benefit coverage – in line with those for benefit generosity – have so far not been successful.

The analysis of relative unemployment ratios for youth and adult populations, based on labour force survey data, shows that the labour market for workers below the age of 25 is more volatile, especially for the 20-24 group. Cross-country differences are, nevertheless, remarkable despite the greater risks generally experienced by young people and the widespread impact of the crisis. The ratios of youth and teen unemployment, respectively, to prime-age unemployment tends to be stable and in some cases has improved over time rather than worsening. In other words, over the last 15 years the relative disadvantage of young people in the 20-24 group compared to prime-age individuals has been declining in Europe (based on the ratio of unemployment ratios for the two groups) (see O'Reilly et al. 2015 for an explanation of different measures). That trend has not changed with the crisis.

In the final section we conclude that youth is not a homogenous group either within or across countries. Our results suggest that methodologically the analysis of the youth labour market should aim to capture the observed heterogeneity of the youth category and adopt a range of metrics to analyse complex trends. Furthermore additional work is required to improve the reliability of certain institutional-level data, for example the information on benefit coverage rates in a cross-national perspective and in relation to young people.

Among our country clusters we find variations in a range of outcomes over the crisis period between age groups within the wider youth category and between young women and men. Policy makers thus need to adopt a holistic view of the challenges facing young people. The trends in the proportions of young people not in employment education or training (NEET) are a particular concern for policy makers and consideration of the gender dimension to these trends is required, Policy needs to address the security deficit for young people since they are not only more prone to falling out of employment or failing to regain employment, but are also less likely to have access to income security provided by unemployment benefits -- accumulating negative outcomes of flexibility with at best uneven coverage of security measures.

Key words: Flexicurity; flexibility; security; youth


Abbreviations AHC Agglomerative Hierarchical Clustering ALMP Active Labour Market Policy/Policies AT Austria BE Belgium BG Bulgaria CH Switzerland CY Cyprus CZ Czech Republic DE Germany DK Denmark EE Estonia EL Greece EPL Employment Protection Legislation ES Spain ESF European Social Fund EU European Union EU-LFS European Union Labour Force Survey EU-SILC European Union Survey on Income and Living Conditions FA Factor Analysis FI Finland FR France FSCA Flexible and Secure Contractual Agreements GDP Gross Domestic Product HU Hungary IE Ireland IS Iceland ISCED International Standard Classification of Education IT Italy LLL Life-Long Learning LT Lithuania LU Luxembourg LV Latvia MT Malta NEET Not in Employment, Education or Training NL Netherlands NO Norway OECD Organisation for Economic Cooperation and Development PL Poland PT Portugal RO Romania

6 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith SE Sweden SI Slovenia SK Slovakia SOC Standard Occupational Classification SSS Social Security System STW School to Work UK United Kingdom VET Vocational Education and Training


Table of contents Table of contents ............................................................................................................................... 7

1. Introduction ................................................................................................................................... 8

Raul Eamets, Katrin Humal and Mark Smith ............................................................................. 8 1.1 Policy background ................................................................................................................... 8

1.2 Aims and Organisation of this paper ..................................................................................... 10

2. Mapping countries for flexibility and security during the crisis framework: empirical evidence ... 11

Raul Eamets, Kariappa Bheemaiah, Katrin Humal and Mark Smith ....................................... 11 2.1. Factor Analysis of flexibility-security Indicators .................................................................... 12

2.1.1. Institutional settings ....................................................................................................... 12 2.1.2. Outcomes ...................................................................................................................... 14

2.2. Cluster analysis of flexicurity indicators ................................................................................ 15

2.2.1 Institutional indicators ..................................................................................................... 15 2.2.2 Outcome Indicators ........................................................................................................ 17

3. Tracing flexibility and income security for youth during the economic crisis ............................... 20

Janine Leschke and Mairead Finn .......................................................................................... 20 4. Comparative analysis of volatility and long-term patterns in youth unemployment ..................... 23

Ilaria Maselli and Miroslav Beblavý ......................................................................................... 23 4.1. Ratios versus rates ............................................................................................................... 23

4.2. Teens (15-19) are different from youths (20-24) .................................................................. 23

4.3 Unemployment, employment and labour market participation .............................................. 24

5. Synthesis and Conclusions ......................................................................................................... 26

Raul Eamets and Mark Smith .................................................................................................. 26 Bibliography .................................................................................................................................... 29

Recent titles in this series ............................................................................................................... 32

Research Partners .......................................................................................................................... 35

Advisory Groups .............................................................................................................................. 36

8 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith

1. Introduction Raul Eamets, Katrin Humal and Mark Smith

1.1 Policy background The term Flexicurity was a dominant theme at the European and national level in the first decade of this century and although its use has declined in recent years the balance of flexibility and security of labour market participants remains a central factor in determining labour market outcomes for young people. In particular it is the tension between providing the flexibility to create jobs while avoiding the negative outcomes of low levels of security which remains important for the levels of objective and subjective security young people face on the labour market (van Oorschot and Chung 2015), the dynamics of segmentation for youth labour markets and the barriers for their integration (Leschke 2012). It is with this broader concept of “Flexicurity” that we approach the term in this report where we map flexibility-security outcomes at the national and individual level during the crisis. Perhaps the most used definition of Flexicurity is that proposed by two of the main and original proponents of Flexicurity, Wilthagen and Tros, who state that Flexicurity;

“is (1) a degree of job, employment, income and ‘combination’ security that facilitates the labour market careers and biographies of workers with a relatively weak position and allows for enduring and high quality labour market participation and social inclusion, while at the same time providing (2) a degree of numerical (both external and internal), functional and wage flexibility that allows for labour markets’ (and individual companies’) timely and adequate adjustment to changing conditions in order to enhance competitiveness and productivity” (Wilthagen and Tros 2004:170).

This definition captures the essence of the concept and avoids an over emphasis on synchronicity and deliberateness of policy which evolve in a variety of national contexts and may be the result of various compromises and negotiations and indeed include tensions and contradictions (see Philips and Eamets 2007). However, criticisms of the Flexicurity concept extend further. At the root of many of the criticisms of Flexicurity are the ambiguities and vagueness around the concept and what exactly Flexicurity policy includes (Leschke et al 2006; Calmfors 2007; Auer 2010). This is perhaps one of the reasons the concept was picked up so easily at the policy level across a wide variety of stakeholders. The lack of clarity in definition, however, perhaps also explains how policies resulting in an overemphasis of (external) flexibility and employability with little emphasis on job and income security have developed (for example Tangian 2007; Burroni and Keune 2011; Heyes 2011). The criticism has also come from other perspectives too. The specific national origins of the Danish and Dutch models that were used to promote Flexicurity are not necessarily transferable to other contexts (Bredgaard et al. 2005). There was also a failure to recognise the gendered processes on the labour market and uneven experience of flexibility and security (Rubery et al 1998; Plantenga et al. 2007; Jepsen 2008; Smith and Fagan 2008). In addition the economic crisis put the Flexicurity concept further under pressure by exposing the limited implementation of policies and the uneven security afforded to different labour market groups (Heyes 2011; Ibsen 2011). Tangian (2007) indicates that flexibilisation has resulted in an increase in unemployment as well as an increase in the share of atypical (part-time or fixed-term employment) and self-employment. As a result, the social protection schemes that are typically designed for full-time permanent workers do not cater sufficiently for the needs of more flexible workers (Vielle & Walthery, 2003; Leschke, 2012). In most European countries workers on non-standard contracts have more limited access to unemployment benefits than workers on standard employment contracts, which exacerbates the position of vulnerable labour market groups that are often disproportionally engaged on non-standard employment contracts, such as young people, women and people with a low education level. As a result, unemployment coverage is often lower for women and young people compared to core-age


men (Leschke, 2012). The problem of a segmented labour market, where employees on non-standard contracts are in a disadvantaged position in terms of security, has been acknowledged on EU level as well as on country level in a number of member states It is, nevertheless, important to recognise that the acceptance of the Flexicurity approach did represent something of a shift away from more deregulatory stances adopted by bodies such as the EC and OECD to one that acknowledged the importance of security on the labour market (Villa 2007). More recently the unemployment crisis, for both core-age and young people, has brought into question the previous strong focus on external flexibility. A central tenant of the European Employment Strategy (2000-2010) had been the push for more jobs with a waning focus on job quality as quantitative targets and raising employment rates became key policy targets (Smith et al. 2009). A reappraisal resulting from the economic and financial crisis has led to some shift in focus to the role of job security, adequateness of income security (unemployment benefits) and the role of internal flexibility (success of short-time working measures and working-time accounts) in buffering the crisis (Leschke 2013; see also chapter 3 below). The Employment Package launched in April 2012 draws specific attention to e.g. encouraging companies' internal flexibility to protect jobs in crisis times, as well as to reducing the labour market segmentation between those in precarious employment and those on more stable employment (European Commission, 2015). Young people are particularly at risk from the asymmetric application of Flexicurity policies. Youth tends to accumulate negative Flexicurity outcomes due to their greater exposure to numerical external and fewer opportunities to benefit from internal flexibility than the comparable adult population (Madsen et al. 2013). At the same time young people also have less job and income security due to the weaker position in terms of seniority. The crisis, with youth being particularly hard hit, has further emphasised these negative outcomes (Choudhry et al. 2012), and as a response on the EU level, the Employment Package includes a specific Youth Package to tackle unacceptable levels of youth unemployment and social exclusion by giving young people offers of jobs, education and training (European Commission, 2012). However, national policy responses have had potentially contradictory effects on young people with some more inclusive unemployment benefits, the extension of short-time working on the one hand and austerity measures targeting social protection and income transfers on the other. Criticism around the ambiguous definition of Flexicurity is likely to extend to a mixed picture in terms of implementation and performance when we look at the labour market outcomes for European countries over recent years. It is for these reasons that the subsequent chapters place a heavy emphasis on analysing flexibility-security components rather than engaging in a debate about the overall concept. This allows us to both explore potential tensions and contradictions but avoid conceptual fuzziness, and negative publicity in some quarters, associated with the concept. We underline that the changing policy and labour market context for young people makes the study of policies that can be broadly labelled “Flexicurity” a relevant and timely activity regardless of the usage of the term itself. Indeed young people have always been at the nexus of this tension between flexibility and security on the labour market (for example Ashton et al. 1988). In Europe today young people find themselves facing the structural challenges that they have always faced as new labour market participants combined with the difficulties created by the repercussions of the economic and financial crisis and ongoing austerity measures.


1.2 Aims and Organisation of this paper The aim of this paper is to explore the inter-linkages between flexibility and security from multiple perspectives in order to uncover both the institutional configurations that contribute to different outcomes for young people and also the impact of the economic and financial crisis and resulting policy changes. In undertaking these analyses we aim to outline how the crisis has affected the social protection within different policy regimes, identify which policy regimes have contributed to improving/deteriorating youth outcomes with regard to flexibility and security. Furthermore we underline the extent to which groups of countries with apparently similar institutional configurations have (dis)similar outcomes. Our aim is also to provide a framework, and catalyst for, subsequent analyses in WP10 of the STYLE project and their research on flexibility-security nexus for young people more widely (www.style-research.eu/project/work-packages/wp10-flexicurity).

This report is made up of five major sections. Firstly, this introduction sets out the main aims of the report and critically reviews the concept of Flexicurity and its application to young people. This opening section provides a framework by which more in-depth analysis of flexibility and security is developed in subsequent chapters. Chapter 2 presents a comprehensive mapping of flexibility and security measures and outcomes for European labour markets. Building on previous mapping exercises this contribution uses two alternative methodologies to map flexibility and security for both youth and the total labour market during the pre- and post-crisis periods. Chapter 3 tracks the changes in policies towards income security during the crisis, on the one hand, and internal and external flexibility on the other. This contribution is particularly useful in highlighting the trends in policy making and the uneven impacts upon young people across the EU. Chapter 4 focuses on one specific outcome of insecurity by examining in detail the ratio of youth and core-age unemployment rates during the crisis. This measure is particularly useful for assessing the performance of EU labour markets vis-à-vis young people. Finally chapter 5 presents a synthesis of the results and considers the implications for future analysis of flexibility and security policy and flexibility-security outcomes for young people.


2. Mapping countries for flexibility and security during the crisis framework: empirical evidence Raul Eamets, Kariappa Bheemaiah, Katrin Humal and Mark Smith

This section looks at institutional configurations and related outcomes for flexibility-security across EU member states during the period 2007–2011 in order to investigate the impact of the first phase of the crisis, compared to the later period. The starting point is the EU approach to the four aspects of flexicurity:

• Flexible and Secure Contractual Agreements (FSCA), • Active Labour Market Policies (ALMP), • Life-Long Learning (LLL), • and Social Security Systems (SSS)).

Two alternative analyses are presented for mapping the institutional settings: one focuses on locating the countries on a flexibility-security plane at two points in time (2007 and 2011), while the other presents analyses how groups of similar countries have evolved over the whole period. Both then look at a number of outcomes in the light of the institutional configurations. The methods are similar to those used in previous flexicurity research, namely factor analysis (see e.g. European Commission, 2006; Philips & Eamets, 2007), cluster analysis (see e.g. European Commission, 2006; Philips & Eamets, 2007; Auer, 2010; Chung, 2012) and comparative analysis (see e.g. Wilthagen, 2004; Sperber, 2005; Tangian, 2007; Maselli, 2010; Auer, 2010; Muffels et al., 2010; van Vliet & Nijboer, 2012; Chung, Bekker & Houwing, 2012). As with many other methods, the results of factor and cluster analyses are somewhat dependent on the selection of included variables; however, they do provide a useful overview of the flexibility-security profiles of EU countries. In addition, alternative variable combinations have been used for robustness checks.

The selection of indicators (obtained from Eurostat and OECD databases) draws heavily on those used in previous empirical studies (see Table 1). However, in seeking to develop a cross-sectional comparison over time and explore the impact upon the youth labour market, further constraints became part of the selection criteria. For instance, the OECD Employment Legislation (EPL) index is missing for several EU members and thus excludes a number of countries from the analysis. Additionally, in some cases data are missing about expenditure on education or Active Labour Market Policies (ALMP). The choice of outcome measures is additionally limited by data availability for age groups (e.g. there are data gaps in the youth long-term unemployment rate or their participation in ALMP measures).

12 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith Table 1. Flexicurity indicators used in earlier studies

Aspect Indicator Used by

ALMP Expenditure on ALMP Andersen (2011); Chung (2012); Van Vliet & Nijboer (2012); Maselli (2010)

ALMP/SSS Expenditure on Labour Market Policies (LMP) EC (2006); Auer (2010)

SSS

Expenditure on Passive Labour Market Policies (PLMP)/UIB Chung (2012); Sperber (2005)

Expenditure on social protection Auer (2010); Philips & Eamets (2007)

GINI coefficient Philips & Eamets (2007); EC (2006)

Unemployment Insurance Benefit (UIB) replacement rates

Philips & Eamets (2007); Andersen (2011); Chung (2012); Maselli (2010); Van Vliet & Nijboer (2012)

Poverty risk Philips & Eamets (2007); Chung (2012); Maselli (2010); EC (2006)

FSCA

Employment Protection Legislation (EPL) index

EC (2006); Andersen (2011); Bertozzi & Bonoli (2009); Chung (2012); Maselli (2010); Sperber (2005); EC (2006); Tangian (2007); Auer (2010); Van Vliet & Nijboer (2012)

Unemployment rate Chung (2012); EC (2006); Sperber (2005); Philips & Eamets (2007); Andersen (2011)

Long-term unemployment rate Maselli (2010); Philips & Eamets (2007); Andersen (2011); Chung (2012); EC (2006)

Job tenure Maselli (2010); Eurofound (2007); EC (2006); Auer (2010); Chung (2012)

Share of atypical employment (temporary, part-time and/or self-employment)

Van Vliet & Nijboer (2012); Chung (2012); Wilthagen (2004); Philips & Eamets (2007)

Employment rate Philips & Eamets (2007); Chung (2012); EC (2006) Transitions from unemployment to employment EC (2006); Wilthagen (2004)

LLL Participation in education and training in 4 previous weeks

EC (2006); Maselli (2010); Eurofound (2007); Philips & Eamets (2007); Chung (2012)

Note: SSS – Social Security Systems, FSCA – Flexible and Secure Contractual Agreements, ALMP – Active Labour Market Policies, LLL – Life-Long Learning

2.1. Factor Analysis of flexibility-security Indicators

This section maps institutional settings of flexibility and security in 18 EU countries. First, factor analysis1 is used to reduce the dimensions of the data and allow for a two-dimensional representation of flexibility and security settings. Then the factor scores are used in cluster analysis2 to group countries with similar settings. In addition, outcomes such as unemployment and at-risk-of-poverty rates are compared. The analysis looks at two time points, 2007 (before the crisis) and 2011 (after the first phase of the crisis), as well as two age groups in terms of outcomes – youth (15–24) and prime-age (25+)3. It appears that countries with similar institutional settings do not necessarily have similar outcomes, and that the patterns for youth and prime-age workers are different.

2.1.1. Institutional settings The factor analysis resulted in two factors that express institutional settings. Factor 1 characterises flexibility, consisting of the EPL index for regular contracts (individual and collective dismissals) and the EPL index for temporary contracts. Factor 2 characterises security, consisting of ALMP expenditure per unemployed capita, expenditure on education per student (all levels of education), social protection expenditure on unemployment per unemployed capita, and the overall average of

1 Principal component analysis with oblique factor rotation was used. 2 Hierarchical cluster analysis (complete linkage clustering based on Euclidean distance) was used. 3 Depending on data availability, the upper age limit of the prime-age group is either 54, 64 or 74 years.


net replacement rates over 60 months of unemployment.4 The relative position of the countries in terms of flexibility and security, measured in standard deviations from the mean of the respective year, can be seen in Figure 15. Figure 1. Institutional settings of flexibility and security in 18 EU countries, 2007 and 2011

Note. The variables have been standardised so the graph represents the countries’ relative positions with regard to the average of the respective year. Source: authors’ calculations based on OECD and Eurostat data.

Between 2007 and 2011 some countries, such as Finland, the Netherlands, Italy, Slovenia, the Czech Republic, Poland and Hungary, changed their relative position very little. For others, the change was more noticeable. Compared to the average levels in each year, Denmark became relatively more flexible and less secure; Portugal, Estonia and Sweden more flexible; Slovakia and France less flexible; Belgium less flexible and more secure; Germany and Austria more secure; and Spain and Ireland less secure. It can be seen that if in 2007 Denmark was a clear leader in the security

4 The factors remained the same, regardless of using the expenditure variables as per capita or as % of GDP. Similarly, the results were not affected by using total social protection expenditure instead of social protection expenditure on unemployment only. However, replacing the Net Replacement Rates (NRR) over 60 months with the NRR of the initial phase of unemployment yielded different factors, where one was mostly related to the three expenditure measures, while the other to the two EPL measures and the NRR measure. This might indicate that the long-term NRR is a more appropriate measure to investigate the general levels of flexibility and security. 5 The input indicators of the factor analysis have different measurement units and scales so were standardised before applying factor analysis. Therefore, the movements on the graph are only relative, taking into account the positions of all other countries in the respective year. The figure does not express absolute changes or even their directions. Note that the flexibility factor scores have been inverted on the graph so that flexibility increases but the actual factor scores based on EPL indices decrease from left to right.

AT07 BE07

CZ07

DK07

EE07

FI07 FR07

DE07

HU07

IE07

IT07

NL07

PL07

PT07

SK07

SI07

ES07

SE07

AT11

BE11

CZ11

DK11

EE11

FI11

FR11 DE11

HU11

IE11

IT11

NL11

PL11

PT11

SK11

SI11

ES11

SE11

-‐2

-‐1.5

-‐1

-‐0.5

0

0.5

1

1.5

2

2.5

-‐2.5 -‐2 -‐1.5 -‐1 -‐0.5 0 0.5 1 1.5 2

Security

Flexibility

2007

2011

14 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith dimension then in 2011 Belgium and Austria had achieved similar levels of security. Also, Ireland’s long lead in the flexibility dimension in 2007 had decreased by 2011, with Denmark having a similar level of flexibility.

As for the absolute changes in the input variables, we note considerable decreases occurred in Estonia and Portugal in the EPL index for regular employment, and in Portugal, Spain and Sweden for temporary employment. The only country which notably tightened EPL was Slovakia where there was an increase in the EPL index for temporary employment.

As for security, the Net Replacement Rates (NRR) measure changed very little for all countries except Denmark, where it decreased considerably. Educational expenditure did not generally change much either (although increased a little in many countries). Unemployment-related social protection expenditure increased considerably in Austria, Germany and especially Belgium, while it decreased in Spain, Denmark and Ireland. Belgium, Austria and Germany also increased in ALMP expenditure, while there were notable decreases in Spain, the Netherlands and Italy. Even larger decreases occurred in Denmark and, most of all, Ireland. Slovakia and Estonia can be highlighted as countries with very large relative changes considering their starting levels – increasing ALMP expenditure 1.7 and 2 times, respectively.

Using the factor scores as an input for cluster analysis yielded mostly consistent results with earlier studies (e.g. European Commission, 2006; Philips & Eamets, 2007; Auer, 2010; Chung, 2012). Nordic countries (Denmark, Sweden, Finland), the Netherlands and Ireland belong to clusters with higher levels of flexibility and security; but CEE countries (the Czech Republic, Estonia, Hungary, Poland, Slovakia, Slovenia) are in groups with relatively low social security, while Southern European (Portugal, Spain, Italy) and also Continental European countries (France, Belgium) are in groups with relatively low flexibility (see Figure 1 above). Earlier studies have found Germany and Austria to be closer to France and Belgium, while in this study they appear more similar to the Nordic countries. It is possible that the flexibility-security levels of Germany and Austria have converged with the Nordic countries, but the differences might also arise from the fact that studies use a different set of input variables.

2.1.2. Outcomes The two countries with highest levels of security and flexibility, Denmark and Ireland, both had the lowest youth unemployment rates as well as low NEET rates in 2007. However, in 2011 Denmark was still one of the best performers while both rates had sharply increased in Ireland, being among the worst in the EU. This supports previous suggestions that flexibility-security performance in downturns might not be the same as in economically good times and may vary for different population groups (e.g. Auer, 2010; Andersen, 2011).

At the same time, the Netherlands and Austria, both with relatively high levels of flexibility and security as well, had low youth unemployment both in both 2007 and 2011. The Netherlands had consistently low unemployment rates for prime-age as well, while for example Slovakia consistently had some of the highest unemployment rates for both age groups. Theoretically, strict employment protection has been linked with higher levels of unemployment, but in 2007 Slovakia was in fact relatively more flexible than the Netherlands, while in 2011 it was the other way around. However, the Netherlands had much higher security levels, including expenditure on education and ALMPs, indicating the important role these measures play in addressing unemployment for both youth and older age groups.

Prime-age at risk of poverty and in-work at risk of poverty rates tended to be lower at higher levels of flexibility and security but the same pattern does not appear for youth. Youth at risk of poverty rates were lowest in a range of countries including Slovenia, Austria, the Czech Republic, Slovakia, and in 2011 also Belgium; while highest in Denmark, Sweden, Italy, and in 2011 also Spain. At risk of poverty rate is a household-based measure so the values are higher when young people tend to move away from their parental home early in the life course – in the Nordic countries and the Netherlands where this is the case, the at-risk-of-poverty rate is 2–3 times higher for youth than for prime-age (in other countries usually 1–2 times higher). However, youth in Italy and Spain tend to live


with their parents for longer and yet the youth rates were very high, so this cannot be the only explanation (see Berloffa et al. 2015).

Youth in-work at risk of poverty rates tended to be among the lowest in CEE countries (except Poland), but highest in Denmark and Sweden. Also, the ratio of youth to prime-age rates was notably higher in Nordic countries than elsewhere; while in a number of countries before the crisis, the rate for youth was even lower than for the prime-age group. At the same time, severe material deprivation rates were higher for youth than for prime-age almost everywhere, and for both age groups tended to be highest among CEE countries (with medium flexibility and low levels of security). They were lowest in the Nordic and Continental European countries (with high levels of flexibility and security). The poverty measures are relative to the income level in the country, so working youth in CEE might have similar income levels as older age groups, but the overall material deprivation rates show that youth in CEE are still worse off compared with prime-age persons as well as youth in other countries (e.g. the Nordics).

2.2. Cluster analysis of flexicurity indicators

2.2.1 Institutional indicators To understand the changes in European countries with respect to flexibility and security related measures at the institutional level, a further analysis of the countries was carried out using Agglomerative Hierarchical Clustering (AHC) techniques on national-level data for 21 EU member states6, for each year from 2007 to 2011. The list of the variables used can be seen in Table 2. The AHC helps provide a more detailed understanding of how the countries can be grouped and enables us to trace their evolution with respect to the indicators. AHC7 also permits the identification of a structure among a defined set of variables and identify country groups on different levels that share common characteristics with maximum intra-homogeneity and maximum inter-heterogeneity.

Using this method, we were able to obtain robust results for the crisis period and, it was observed that from 2007 to 2011, the EU countries clustered into three distinct clusters. While most countries consistently found themselves in the same cluster groups for each year between 2007 and 2011, certain countries changed their cluster positions over this time. In order to reflect these groupings we divide the 21 EU countries into four clusters groups.

The first three clusters consist of the countries with consistent membership in their respective cluster groups from 2007 to 2011, while the fourth cluster (seen below as the “unsettled” cluster) is made up of those countries that had oscillating positions across clusters during this time period: § Cluster 1: Austria, Belgium, Finland, France & Greece § Cluster 2: Czech Republic, Estonia, Hungary, Poland & Slovakia. § Cluster 3: Denmark, Germany, Netherlands & Sweden. § Unsettled Cluster: Ireland, Italy, Luxembourg, Portugal, Slovenia, Spain, United Kingdom.

6 Owing to issues of missing data, analysis cannot be carried out on all EU member countries and we are restricted to contain our analysis up to 2011. 7 Ward’s method was selected as the similarity is calculated as the Sum of Squares between clusters summed over all their variables, allowing for the creation of clusters with minimal intra-variation.

16 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith Table 2. Institutional Indicators and Results from the AHC Analysis8

Aspect Indicators Observations per Cluster

Active Labour Market Policies9

II 1: Total expenditure on ALMP as % of GDP

Expenditure on ALMPs showed very little change over this period for all 4 clusters, with the exception of Cluster 1, which showed an overall increase in spending. Denmark and Ireland also showed similar increases.

II 2: Total expenditure on Public employment services (PES) and administration as % of GDP

While most countries showed very little change over this period, the UK reduced its expenditure by 0.4pp. Other countries in Clusters 2 & 3 showed small increments.

II 3: Expenditure on Training as % of GDP

The general trend witnessed across all 4 clusters is one of increased spending in Training, with the exception of Sweden, Luxembourg, Poland & Italy. Then mean values increased by 0.15pp on average.

Social Security Systems

II 4: Expenditure on social protection as % of GDP

All countries showed significant increases across this variable. Ireland stands out with an increase of 12.2pp compared to an overall average of 3.4pp.

II 5: Total expenditure on unemployment as % of GDP10

Spending on unemployment also see’s similar increases across almost all countries, with the exception of Germany which showed a 02pp reduction. Ireland once again shows the largest spending increase of 2.3pp in this period.

Flexible and Secure

Contractual Agreements

II 6: EPL- Individual & Collective Dismissals - Regular Contacts11

While employee protection for regular contracts remains relatively stable, employees with temporary contracts do not share the same security with the average value of II 7 increases in dismissals of persons having temporary contracts increasing by 0.4pp.

II7: EPL- Individual & Collective Dismissals -Temporary Contracts12

Life Long Learning

II 8: Annual expenditure on public and private educational institutions as percentage of GDP, for all levels of education combined

20 of the 21 EU countries in our data set showed an increase in spending on educational institutions, thus increasing overall expenditure by 0.5pp. , The average spending on educational institutions has seen an average increase of 0.5pp (with the exception of Hungary).

II 9: Financial aid to pupils and students as percentage of total public expenditure on education, for all levels of education combined

While countries in Clusters 1 and 3 show almost no increases towards financial aid, the countries on the 4th cluster show an overall increase. The UK showed a noteworthy increase of 3.2pp, while Denmark reduced its aid to students by 1.9pp.

Between 2007 and 2011, the general trend that is seen across all clusters via these variables, is one of increased passive security and less flexibility. Clusters 1 and 3 shows increases in LMP spending (II 1), but little was spent on measures towards employment services (II 2). Although there is an overall increase in spending for education and training (II 3 and II 8), financial aid to students (II 9) has fallen in countries of cluster 1 and 3. The Unsettled cluster also shows divergent results for II 9.

Thus these trends are polarized across contexts. While EU countries continue to provide employment security to those who possess regular contracts - potentially reducing flexibility - employees with

8 In this table, II indicates ‘Institutional Indicator’. 9 These indicators were also tested on the basis of labour market participants instead of GDP which varies across the phases of the crisis. However the results were very similar and since GDP based data provided a more complete list of countries the former was kept. 10 Rerunning these analyses based on the number of unemployed for 2010-12 highlighted that there was an overall increase in spending in cluster 1, an overall reduction in spending in cluster 2 (with the exception of Slovakia and Estonia) and in cluster 3 Denmark and Sweden reduced their spending while the others increased. Again the unsettled cluster shows mixed results. 11 Version 3 of this indicator is the weighted sum of sub-indicators concerning the regulations for individual dismissals (weight of 5/7) and additional provisions for collective dismissals (2/7). It incorporates 13 detailed data items (http://www.oecd.org/els/emp/EPL-Methodology.pdf). 12 Version 3 of the indicator for temporary employment measures the strictness of regulation on the use of fixed-term and temporary work agency contracts. It incorporates 8 data items (http://www.oecd.org/els/emp/EPL-Methodology.pdf).


temporary contracts do not share the same sense of employment stability thus increasing flexibility and insecurity. At the same time, passive labour market policies have taken precedence to active policies such as entrepreneurship investment. Although there has been an increase in investment towards training and education in general, we notice a lack of investment towards job creation, which suggests that an increasing number of young people are remaining in education rather than entering a difficult job market. The focus on passive measures is further amplified with the increased spending in unemployment and social protection. The outcomes of these policy measures especially with regards to youth and from a gender perspective will be analysed in the next section.

2.2.2 Outcome Indicators Having grouped the 21 EU countries into different clusters over the crisis according to the institutional indicators, in this section we aim to explore if the outcomes have been similar to different demographic groups by individually analysing the outcome variables, listed in Table 3. For the sake of simplicity and to measure the similarity of the results across the countries, the values of the outcome variables were analysed by clusters defined in the previous section. The changing mean and standard deviation values of each variable for each cluster were then traced over time, from 2007 to 2013 (refer to separate Appendices), in order to gauge the evolving impact of institutional policies on young people in the EU.

Table 3. Outcome Indicators and Results from the AHC Analysis13 Aspect Indicators Observations per Cluster

Active Labour Market Policies

Out 1: Number of participants as a percentage of labour force: Active Measures14

Between 2007 and 2013, while there has been increased participation in training (notably in Cluster 3), there has been a drop in job creation and start-up incentives. As a result, the mean values show small changes over this period.

Out 2: Number of participants as a percentage of labour force: Passive Measures15

There has been an increase in Out-of-Work-income participation but a decline in measures such as early retirement. This indicates that the labour market policies have not necessarily been conducive to employment over this period and there has been a reduction in job security in general.

Social Security Systems

Out 3: People at risk of poverty or social exclusion by age and sex

Analysis of this indicator reveals that the risk of poverty for the youth population has steadily increased over the period. While the overall mean values show small changes, young men and women between the ages of 16-24 have been adversely affected. In the Unsettled cluster for example, the risk has increased by 10pp for men and women.

Out 4, 5 & 6: Long-term unemployment in terms of % of active population

While Cluster 3 exhibits relatively stable figures, the mean value of the countries in the Unsettled cluster shows a sharp increase in unemployment both for men and women. This trend is greater for the very long-term unemployment. In general, the number of long-term unemployed has increased steadily over this period .

Out 7, 8 & 9: Young people not in employment and not in any education and training (15-29 yrs.)

The mean values of the countries in cluster 1 and 2 show the highest increases over this time period. More specifically the 20-24 male population in cluster 1 and the 25-29 female population in Cluster 2 show relatively high values in comparison to the other clusters. In the Unsettled cluster we also see a 9pp increase in number of male NEETs (20-24) who are seeking employment.

13 In this table, “Out” indicates ‘Outcome Indicator/Variable’. The Output indicators measure the impact of the Institutional Indicators on the country’s population. As with the Institutional Indicators, the Outcome Indicators are organized as per the four aspects of Flexicurity. 14 Includes (1) PES & administration; (2) Training; (3) Employment incentives; (4) Supported employment and rehabilitation; (5) Direct job creation and (6) Start-up incentives. 15 Includes (1) Out-of-work income maintenance and support and (2) Early retirement.


Across all clusters, irrespective of age or sex, we also see a decline in the number of NEETs.

Out 10, 11 & 12: In-work (employed persons) at-risk-of-poverty rate for young people

Young people have seen a steady increase in the risk of remaining poor, even when employed. Cluster 2 and 3 shows erratic trend lines during this period but culminate with high risk values especially for the 15-19 population. The 25-29 population has also suffered an increased risk rate which could explain why a there been an increase in the number of participants in tertiary level education. Men and women have both been affected although for Cluster 3 for the female 25-29 population we see a decline in the mean values from 6.9 to 5.5.

Out 13: Early leavers from education and training (18-24yrs)

An increasing number of young people stay in education rather than working thus we see a steady decline across all clusters and for both sexes with regards to this outcome. This is clear within the Unsettled cluster where the mean value for men dropped from 22.4 to 15.4 and from 15.1 to 10.3. More men leave education early than women.

Life Long Learning

Out 14, 15 & 16: Participation rate in education and training

We see an overall increase in participation in education across all clusters and for both sexes during this period.

Out 17: Activity rates of young people not in education and training- Education level (0-4)

In cluster 1 we see the mean value for activity rates decrease by 3.7pp for men and 2.1pp for women. However, the Unsettled cluster shows a 5.4pp decrease for men and a 7pp decrease for women.

Out 18: Activity rates of young people not in education and training- Education level (3- 8)

A similar albeit slightly varied result is seen for this outcome. While activity rates decreased by 3 pp for men and 2.8pp for women in Unsettled Cluster, we also notice a 2.5pp decrease for women for in cluster 2.

Flexible and Secure

Contractual Agreements

Out 19: Involuntary part-time employment as percentage of the total part-time employment for young people by sex and age-15 to 29 years

An increasing number of young people accepted part-time employment between 2007 and 2011. While the mean values of the countries in cluster 2 saw a 7.7pp increase, the Unsettled cluster saw a 12.3pp increase as well. Cluster 3 however showed a 0.7pp decrease.

Out 20 & 21: Student – Labour Transitions in terms of employment security.

Both changing mean values of both these indicators show similar results. Across all cluster there is a general increase in the transition jobs offering less security. The most extreme values are seen again with the Unsettled cluster that show a 10.8pp increase for men and 7.9pp increase for women. Cluster 3 provides an exception providing transitions with greater security for both men and women.

Out 22, 23 & 24: Young temporary employees as percentage of the total number of employees.

A similar increase is seen in the mean values of temporary employees and part time workers in all clusters irrespective of age or sex although from different starting points.

Out 25: Part-time employment as percentage of the total employment for people aged 15-29

From the analysis of the 25 outcome variables we observe that young people have been adversely affected in terms of employment and income since the advent of the crisis. There is an overall drop in employment security (Out 2, 4, 5 & 6) especially in countries belonging to the Unsettled cluster, although less pronounced for countries in Cluster 3, which is also the only cluster where the situation is slightly more secure for women. However, it is also the cluster that shows the highest values with regards to the risk of young people remaining poor in spite of finding employment, especially for the 20-29 female population (Outcomes 10 to 12). In addition the situation for women NEETs deteriorated further in Cluster 1 and in the Unsettled cluster (Out 8 and 9) during this period.

The participation rates in training have increased across all clusters; they are highest in Cluster 3. An increasing number of young people in all countries prefer to stay in education including men, who previously were more prone to leaving the education system to begin working (Out 13). The fact the inactivity rates (Out 17 & 18) have risen across all clusters, especially the Unsettled cluster provides


further evidence of young people staying in education – and out of the labour market - as a result of a lack of opportunities and thus risking being over-qualified for future opportunities.

The result that men and women in the 25-29 age groups are the most affected by low employment security (Out 21 to 25) and are dependent on social aid highlights the risk for young people, since they stand at the precipice of becoming mature workers. Having little access to security will thus delay their progression to beginning an independent life and a family (Mills et al, 2005). The countries in cluster 3 have shown positive results, especially for young women, in terms of education opportunities, employment and risk of poverty. There is also a higher rate of security in these countries. However, across all countries, there has been a growing tendency to stay in education instead of continuing to move towards the workforce as a result of a poor employment environment.


3. Tracing flexibility and income security for youth during the economic crisis Janine Leschke and Mairead Finn

This section draws on the flexicurity matrix (for details see Wilthagen and Tros 2004) to analyse trade-offs, vicious and virtuous relationships between external and internal numerical flexibility as well as income security (Leschke, Schmid, Griga 2007; Schmid 2009). This matrix enables a combination of dimensions of flexibility and security to be explored in a more advanced way than heretofore. Notwithstanding functional equivalents such as lax regulation of permanent contracts (as is the case in the UK), we operationalise external flexibility by looking at temporary employment shares. Internal flexibility is then operationalised as participation in short-time working. Income security is captured through access to (and level of) unemployment benefits. In this way, we examine whether developments during the economic crisis improved or deteriorated the situation of youth on different dimensions of flexibility and security.

External numerical flexibility is very high among youth, particularly the group of youngest youth (15-24 years). The fact that temporary contracts are frequently involuntary, and less frequently used for training and education, indicates that they are beneficial mostly from an employers’ perspective (to screen new employees). Nevertheless, they do allow youth to gain first-hand labour market experience, albeit with limited job security. This situation could be termed a trade-off between flexibility and security. Youths’ over-representation in temporary employment implies that their employment fluctuates more than that of adults, rendering youth more prone to unemployment. Furthermore, on average youth have shorter tenure than adults (due to “last in-first out” rules and an employer preference for employees with greater firm-specific knowledge). It follows that young peoples’ more limited labour market experience, and the predominance of temporary contracts, results in a greater difficulty fulfilling eligibility conditions for access to unemployment benefits (e.g. Leschke 2012). Labour Force Survey (LFS) data demonstrate that youth are also less likely to receive unemployment benefits than adults in almost all countries (especially young youth). This combination of higher contractual flexibility and unemployment, and lower income security during unemployment, can be termed a vicious relationship between flexibility and security.

The initial crisis period, containing stimulus measures, witnessed both an opening up of short-time working measures to new groups of workers and their extension to a larger number of countries (though often temporary). These can be viewed as tools of internal numerical flexibility, which both preserve jobs, and cushion working time reductions to a certain degree, thereby granting income security. According to the LFS data, young workers held a reasonable share of short-time working positions. Where reduced wages due to reduced working-time are compensated in part, this can be seen as a virtuous relationship between flexibility and security. Where they are not compensated a trade-off occurs.

In light of surging youth unemployment – indeed a youth unemployment crisis – in a number of European countries, the limited access of youth to unemployment benefit schemes in many countries has appeared on the international and supranational agenda (e.g. OECD and European Commission focus). This occurred especially in the early period of the crisis. The previous focus on supply-side measures was no longer deemed effective due to the lack of realistic possibilities to bring large numbers of youth back into employment quickly. A number of European countries accordingly improved the situation for young people. More generally, temporary workers also experienced improvements in access to and the generosity of unemployment benefits schemes. This was achieved by relaxing qualifying criteria; offering lump sum or one-off payments; and increasing the amount or duration of benefits. However, already during the first period of the crisis (here defined


2008-2010), reforms to unemployment benefits were not only towards greater generosity. While no countries restricted access to benefits during the stimulus period, a sizable number had already reduced the level or duration of benefits before 2011. In the second crisis period (2011-2014), some countries again tightened qualifying criteria or benefit duration. The general focus in this period seems to have reverted from one on income security to the pre-crisis focus on supply-side measures (for details see table 4).

Table 4. Countries with modifications to unemployment benefit systems or short-time working schemes during the first period of the crisis (2008-2010) and austerity (2011-2014)

(Temporary) modifications Countries 2008-2010 Countries 2011-2014 Relaxing qualifying criteria (eligibility)

Finland, France, Portugal, Latvia, Slovenia, (Slovakia)

Slovenia, Portugal, Italy

Tightened qualifying criteria Czech Republic Lump-sum/one-off payments Greece, France, Italy, Spain Benefits to promote labour market integration***

Austria, Croatia, Estonia, Finland, France, Poland, Romania, Slovakia, Slovenia

Increasing benefit level Belgium, Netherlands, Bulgaria, Czech Republic; Poland

Decreasing benefit level Greece, Ireland, Spain, Latvia Increasing benefit duration Finland, Romania, Latvia, Lithuania Decreasing benefit duration Ireland, Czech Republic, Poland,

France, Denmark Greece

Access of non-standard workers to long-standing short-time working schemes**

Austria, Belgium, Germany, France, Luxembourg

Newly introduced and usually less generous and temporary short-time working schemes

Bulgaria, Czech Republic, Hungary, Lithuania, Latvia, Netherlands, Poland, Slovenia, Slovakia)

*Depending on the country and the reform, the changes affected either all unemployment benefit recipients or certain categories of workers. **Here only schemes are referenced which were previously limited to regular workers. ***Lump sum benefits specifically to cover the costs of returning to work (MISSOC, 2014). Source: own depiction based on various sources.

Using special extracts of the LFS data on access to unemployment benefits, and notwithstanding the limitations of this data (especially compositional effects besides changes in access due to changing eligibility), our analysis reveals an improved situation in coverage for both groups of youth, in the first period of the crisis. This is in line with the institutional changes discussed above. Adults also experienced this in a sizable number of countries. The latest available data, taking into account the austerity period, show that on average across Europe, both younger and older youth are worse off than before the crisis. This is not the case for adults. Accordingly, benefit coverage for youth, which is considerably lower than that of adults, has decreased further in a number of countries (for details see Table 5).

22 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith Table 5. Relative change in access to unemployment insurance and assistance benefits to 2007 in stimulus (2009) and austerity (2013) period age Substantial decrease in

access to UI Substantial increase in access to UI

missing data EU27 (relative level 2013 to 2007)

EU27 pp- change 2013 to 2007

15-24 2009 CY GR, SE, FR, SK, ES, DK, PT, SI, IT

IE, NL, BG, EE, LT, LU, LV, MT

99 -0.1

2013 CY, PT, CZ, PL, ES, AT (2012), HU

RO, DK, GR, IT 81 -3.3

25-29 2009 GR PL, UK, PT, CY, ES, SI, IT, RO

IE, NL, BG, EE, LT, LU, MT

100 0.1

2013 GR, SE, AT (2012) UK, PL, RO, SI, IT

85 -5.4

30-64 2009 LU IT, PT, BG, LV, ES, EE, LT, MT

IE, NL 102 0.9

2013 MT, RO UK, ES, IT 104 1.7 Source: based on Eurostat LFS special extracts. Cut-off points for substantial decrease <75% of 2007 value and for substantial increase >125% of 2007 value. Note: Duration of unemployment 1-2 months. Registered at PES and receiving benefits or assistance as % of all unemployed.

Reliable unemployment benefits of a certain generosity and duration render it possible to search for an adequate job. This facilitates a better match between education and occupation instead of forcing unemployed youth to take the first best option – potentially informal or casual labour. There is also a wider societal effect as such benefits put youth in a situation of independence from their families from which they can consider family formation. Another crucial question is whether there are functional equivalents, such as possibilities for further training or education. This highlights that the context matters. For example, countries such as Spain, Portugal, Cyprus and Slovakia combine very high youth unemployment rates with very low relative benefit coverage rates. By contrast, Denmark’s low rate of unemployment benefit coverage for young people has to be seen in light of a relatively small population of unemployed youth; a high focus on obligatory activation as part of the unemployment benefit schemes; a generous education allowance; and relatively generous social assistance which can act as functional equivalents to unemployment benefit coverage.


4. Comparative analysis of volatility and long-term patterns in youth unemployment

Ilaria Maselli and Miroslav Beblavý

The public debate around the issue of youth unemployment is frequently misled by misinterpretation of the numbers. Our analysis emphasises that the unemployment rate is not necessarily a good indicator for the labour market of those below 25. We also show that youths and teens cannot be looked as a single group. In particular, the labour market of the 20-24 group is more volatile. However, we also show that their relative disadvantage compared to prime age individuals has been declining in Europe. This is our most surprising finding also considering that the trend has not changed with the crisis. We also show that the interplay between education participation levels and labour market opportunities is more complex than conventional wisdom suggests.

4.1. Ratios versus rates In order to compare youth unemployment with that of prime age individuals, the ratio provides more nuanced measure than the traditional unemployment rate and one that is also closer to what is understood as youth unemployment in the public discourse. The ratio is based on the same numerator as the rate (the number of unemployed individuals – that is those that do not have a job and are actively looking for one), but the denominator is different. In the case of the unemployment rate the denominator is the active population that is the sum of workers with and without a job. In the case of the ratio the denominator is the entire population (O’Reilly et al. 2015).

For example, Greece and Finland have very similar unemployment ratio, but entirely different unemployment rates (see Table 6). In Italy, the unemployment rate of teenage workers is 70%, as repeatedly reported in the news at each new release of Eurostat statistics. However, looking at the unemployment ratio is clear that only 5 every 100 teenagers are unemployed, and not 70 as understood in the public discourse. The difference comes from the fact that the unemployed workers are only a small portion of the youth and teenage population which is otherwise either inactive and/or in education.

Table 6. Unemployment rates and ratios compared – 2013q4

15 - 19 24 - 54 Unemployment. Rate Unemployment. Ratio Unemployment. Rate Unemployment. Ratio

Greece 71.4 5.4 27.3 22.8 Spain 75.5 9.4 24.2 21.1 Italy 70.3 5.2 11.6 8.9 Netherlands 14.4 8.8 6.0 5.2 Finland 21.0 5.4 6.8 5.9 Slovakia 63.1 4.4 12.8 11.2 Source: Authors’ elaboration on LFS data

4.2. Teens (15-19) are different from youths (20-24) Another important clarification to be made is that the youngsters’ category is composed of two very heterogeneous subgroups: the youth (20-24) and the teens (15-19). Youth, more than teens, exhibit volatile unemployment rations compared to prime-age workers. Not much is known on the volatility of

24 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith unemployment, despite the fact that economic agents (and economic theories) have a preference for smoothing. Some find that stronger real wage rigidities make unemployment more volatile and that employment protection legislation (EPL) reduces the cyclical response of unemployment, while more rigid wages increase the response of the real economy. 16 Others, instead, argue that higher employment protection legislation increases volatility and that union coverage and wage setting agreements increase the volatility of output.17 Another possible explanation for the comparative difference is that youth unemployment is more volatile when output is also more volatile18.

An interesting piece of information can be traced in the ratio between adult and youth unemployment ratios (RoR). The RoR can be interpreted as a measure of the structural conditions of the labour market when it comes to youth. A high average ratio (above 2), such as the one observed in Italy, Sweden and the UK indicates that younger workers have much harder time than adults to find a job. However, we also observe the RoR over time and surprisingly enough, we find it quite constant in Europe as a whole. If anything it decreases rather than increasing over time, meaning that there is some convergence between the youths’ and adults’ segments of the labour market19, even in countries like Greece, Spain and Italy.

4.3 Unemployment, employment and labour market participation We explore the comparative relationship between activity, employment and unemployment ratios. These are graphically shown in Figure 2 and Table 7, where the key differences between the two young age groups are once more confirmed.

Figure 2. Correlation coefficients between unemployment and employment ratios, based on average values for 1998-2013 (one dot = one country)

Source: Authors’ elaboration on LFS data

16 See for instance: Faccini and Rosazza (2012), Abritti & Weber (2010). 17 Gnocchi and Papa (2011), Rumler and Scharler (2009), Pissarides (2009). 18 Banerji et al. 2014 19 The regression between the mean and time is significant at the 5% level.

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

Unem

ployment ratio

Employment ratio

15-‐19 20-‐24 25-‐54 Linear (15-‐19) Linear (20-‐24) Linear (25-‐54)


For the 20-24 year olds, there is a negative relation between employment and unemployment ratios, similar to the prime age individuals though weaker and with much wider dispersal. In other words, there is a comparative trade-off between employment and unemployment ratios. For teenagers, the reverse applies – there is a mildly positive slope linking employment and unemployment that suggests that the higher unemployment ratios of the teens might be explained by the higher participation on the labour market than is observed in certain countries where there it is more common to combine work and study already at a younger age. This finding underlines that the interplay between education participation levels and labour market opportunities is complex, contrary to what might be suggested by the recent attention to one of the problems of the youth labour market.

Table 7. Correlation coefficients between unemployment ratios and activity/employment rates

Activity ratio

Employment ratio

15-19 0.54* 0.44 20-24 - 0.44 - 0.68* 25-54 - 0.23 - 0.68*

Source: Authors’ elaboration on LFS data

We display in table 7 the correlation coefficients between the unemployment, employment and activity ratios. The star denotes a significance level of at least 5%. As with employment, the correlation between unemployment and activity ratios is negative for youths and prime-age workers and positive for teens.


5. Synthesis and Conclusions

Raul Eamets and Mark Smith

The Flexicurity concept is by definition controversial. It is good label for policy documents but a difficult term if we try to find theoretical bases or model for it. At the core of the concept is the delicate balancing act between flexibility and security that underlines the difficulties in matching labour market security and flexibility simultaneously. Therefore the most common definition of flexicurity stresses security aspects of vulnerable groups (Wilthagen and Tros, 2004). In this report we have adopted a broad definition of the topic in order to analyse policy and outcomes related to flexibility and security rather than being constrained by specific definitions and controversies over the term. Thus we try to use flexicurity term in more neutral form. For example, we discuss finding a balance between social security and labour market flexibility rather than acting as proponents of a particular country model.

One of the challenges of implementing so-called flexicurity policy has been its heterogeneous interpretation and application across countries. In considering the implementation of flexicurity policies, then it seems to be particularly difficult for old Europe (EU15), as social security has been crucial element of the European social model concept. In new member states, on average, relatively low social security was combined with relatively high labour market flexibility. As our analyses show we cannot find unique and one-size-fits-all model of social security and labour market flexibility.

The heterogeneity of flexicurity policy also extends to different groups on the labour market and for this reason the analysis of young people is particularly salient. We know from earlier literature that young people are a particular at-risk group. Young people tend to experience negative aspects of numerical external flexibility and they have fewer opportunities from internal flexibility as they hold weaker positions inside firms due to seniority issues. In this report we present an analysis of the position of young people in relation to flexibility and security from a number of perspectives.

For the flexibility-security mapping exercise we used data from the OECD and Eurostat in two different approaches. One presents the countries’ relative flexibility-security position at two time points, 2007 and 2011, and the analysis other traces the evolution of country groups over the whole period of 2007–2011. The two different methods are complementary and provide us with a possibility for robustness check. In fact with both methods we find quite consistent clustering: East European countries consistently group together, as do Nordic countries, the Netherlands and Germany. A challenge for these analyses was data limitation – missing data and incompleted time series are a typical issue in social sciences. For instance, one of the most frequently exploited indicators of flexicurity analyses, the EPL index, is available only for OECD countries, i.e. not all EU countries. Nevertheless, we construct a detailed dataset for 21 countries with indicators on various measures for flexibility and security and by gender and different youth age groups.

Our aim was to analyse data at least in two time points: before and after the initial effects of the crisis and for comparison we use in addition youth and adult populations as comparison groups. In the selection of indicators we relied on those used in previous studies.

Comparison of selected outcome indicators showed that country groups with similar institutional settings do not necessarily have similar labour market and/or social outcomes for young people. As Chung (2012) previously found countries do not necessarily group in terms of institutions and outcomes and correlations of the two types of measures are often weak. Indeed, this analysis showed that a cluster of countries with similar institutional settings could often include both the best- and the worst-performing countries in terms of outcomes. There are likely to be a range of forces shaping outcomes on the youth labour market leading to within-cluster, and indeed within-country, variations. These results further underline that youth are far from a homogenous group and this heterogeneity


needs to be taken into account in analysing, and attempting to explain, labour market outcomes for young people.

Another finding is that under similar institutional settings, outcomes for different labour market groups vary, such as for youth and prime-age employees but also the elderly, women, the long-term unemployed and temporary employees (see also Maselli (2010) and Leschke (2012)).

The more detailed policy analysis (section 3), making use of Eurostat LFS special extracts and various other sources, shows that it is important to not limit the analysis to youth between 15 and 24 and to not just merge the younger and older youth groups (25-29). We show that older youth are better off than younger youth groups in terms of external – but not internal – numerical flexibility (however we lack detailed and age-specific information on compensation at short-time working). Older youth are also better off with regard to income security. Furthermore, both groups of youth differ from adults in that they do worse on all three dimensions.

Secondly, the policy analysis has emphasised how complex unemployment benefit schemes are; how much they vary across Europe in terms of coverage, generosity, and the availability of secondary schemes; and that they are frequently adjusted (and not always in a strategic way, as seems to have been the case during the first part economic crisis). Thus, comparative analysis of access to benefits is difficult. Attempts to create “simple” indices for benefit coverage – as they exist for benefit generosity (OECD: Benefits and Wages) – have so far been unsuccessful (see Alphametrics 2009). Again the data availability testifies to this complexity. Indeed, as the LFS does not allow a distinction between insurance and assistance benefits, we put a question mark on the reliability of the information on benefit coverage rates in a cross-national perspective and therefore only used relative change within countries in our analysis.

The volatility analysis of relative unemployment rates is the subject of the third pillar of our analysis. Based on LFS data shows that the labour market of workers below the age of 25 is more volatile, especially for the 20-24 group. Furthermore cross-country differences are remarkable despite this common observation and the widespread impact of the crisis. The ratio between youth and teen unemployment, respectively, to the prime-age unemployment tends to be stable or even improving over time rather than worsening. In other words, the relative disadvantage of young people in the 20-24 group, compared to prime age individuals, has been declining in Europe over the last 15 years (based on the ratio of unemployment ratios for the two groups). That trend has not changed with the crisis.

Based on the analyses we can conclude the following.

Methodological issues

- Unemployment ratios complement unemployment rates in order to provide a more accurate picture of the labour market of youths.

- Youth and teens cannot be looked as a single group; we also need to differentiate older (25-30) and younger youth groups (15-19 and 20-24).

- The data for unemployment benefit coverage are very complex. LFS does not allow a distinction between insurance and assistance benefits. We question the reliability of the information on benefit coverage rates in a cross-national perspective.

- Methodologically we find support to looking at institutions and youth labour market outcomes separately since performance in terms of one and the other are not necessarily correlated.

Outcomes

- In general we can conclude that youth tend to accumulate negative outcomes of flexibility (greater exposure to numerical external flexibility). For example higher rates of temporary employment, and in some countries particularly for women.

- Youth is not a homogenous group and within countries and clusters we find variations in outcomes by age groups within the youth category and between young women and men.

- Youth are not only more prone to becoming unemployed, but are also less likely to have access to unemployment benefits.


- A number of European countries improved the security situation of youth before the crisis, but as result of policy changes and austerity measures we see that on average both younger and older youth are worse off after the crisis.

- We observe an increase NEET rates for age cohorts 20-24 and 25-29 for almost all country clusters and both gender groups. Although NEET rates for 15-19 were relatively steady over the time period 2007-2013.

- The stability in teenage NEET rates underlines that mostly those on labour market were hit by recession and the risk of NEET status is concentrated on the older age groups of young people.

- In terms of flexibility, Nordic and Continental European countries, as well as Ireland, appear to have higher levels of flexibility and security than Southern, Central and East European countries.

- There is a growing tendency among young people to stay in education instead of moving into employment. The effect was stronger in Nordic countries, and we did not find any gender differences.

- In spite of the raft of policy making during the crisis, country clusters remain rather stable when we analyse them before, during and towards the end of the initial phase of the crisis.


Bibliography

Abritti, M., Weber, S. (2010), Labor Market Institutions and the business Cycle Unemployment Rigidities vs. Real Wage Rigidities, Working Paper Series, ECB Working Paper Series, No 1183/ April 2010.

Alphametrics (2009) Flexicurity: indicators on the coverage of certain social protection benefits for persons in flexible employment in the EU, study commissioned by the European Commission, VC/2007/0870.

Andersen, T. M. (2011) A Flexicurity Labour Market in the Great Recession: The Case of Denmark. IZA Discussion Paper No. 5710, May 2011

Ashton, D. N., Maguire, M. J. and Spilsbury, M. (1988) 'Restructuring the Youth Labour Market', Labour and Society 13(4): 415-432.

Auer, P. (2010) What's in a Name? The Rise (and Fall?) of Flexicurity. The Journal of Industrial Relations, 52(3), 371-386

Banerji, A., Saksonovs, S., Lin, H., Blavy, R. (2014), Youth Unemployment in Advanced Economies in Europe: Searching for Solutions. IMF Staff Discussion Note, December 2014.

Berloffa, G., M. Filandri, E. Matteazzi, T. Nazio, N. Negri, J. O’Reilly, P. Villa and C. Zuccotti, (2015) ‘Work-poor and work-rich families: Influence on youth labour market outcomes’, STYLE Working Papers, STYLE WP8.1 CROME, University of Brighton, Brighton. http://www.style-research.eu/publications/working-papers/

Bredgaard, T., Larsen, F., Madsen, P. K. (2006), Opportunities and challenges for flexicurity - The Danish example. Transfer: European Review of Labour and Research, 12: 61

Burroni, L., Keune, M. (2011), Flexicurity: A Conceptual Critique. European Journal of Industrial Relations, 17(1), 75–91.

Calmfors, L. (2007), Flexicurity – An Answer or a Question? European Policy Analysis, Nov 2007, 6: 1-5. Swedish Institute for European Policy Studies.

Choudhry, M. T., Marelli, E., Signorelli, M. (2012) Youth unemployment rate and impact of financial crises, International Journal of Manpower, 33 (1), 76–95

Chung, H., Bekker, S., Houwing, H. (2012), Young peole and the post-recession labour market in the context of Europe 2020. Transfer, 18(3), 301–317.

Chung, H. (2012) Measuring Flexicurity: Precautionary Notes, a New Framework, and an Empirical Example. Social Indicators Research, 106, 153–171

European Commission (2006) Employment in Europe 2006. Luxembourg: Office for Official Publications of the European Communities

European Commission (2012) Youth employment: Commission proposes package of measures. Available at: http://ec.europa.eu/social/main.jsp?langId=en&catId=89&newsId=1731 (accessed 21 May 2015)

European Commission (2015) Employment package. Available at: http://ec.europa.eu/social/main.jsp?catId=1039 (accessed 21 May 2015)

Faccini, R., Rosazza, C. (2012), Labour market institutions and unemployment volatility: evidence from OECD countries, Working Paper No. 461 August 2012, Bank of England.

30 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith Gnocchi, S., Papa, E. (2011), Do labor market rigidities matter for business cycles? No 411, Working Papers, Barcelona Graduate School of Economics.

Heyes, J. (2011), Flexicurity, Employment Protection and the Job Crisis. Work, Employment and Society, 25(4), 642–657.

Ibsen, C. L. (2011), Strained Compromises? Danish Flexicurity During Crisis. Nordic Journal of Working Life Studies, 1(1), 45-65.

Jepsen, M. (2008) “Quand la dimension de genre entrera-t-elle réellement en jeu ?” Travail, genre et sociétés n° 19 – Avril 2008

Leschke, J. (2013) La crise économique a-t-elle accentué la segmentation du marché du travail et de la protection sociale? Une analyse des pays de l’UE (2008-2010), Revue Française des Affaires Sociales, No. 4, Special Issue on "Emplois et statuts atypiques : quelles protections sociales?.

Leschke, J. (2012), “Has the economic crisis contributed to more segmentation in labour market and welfare outcomes?” ETUI Working Paper 2012.02, Brussels.

Leschke, J., Schmid, G., Griga, D. (2007) On the Marriage of Flexibility and Security: Lessons from the Hartz-reforms in Germany, in: H. Jørgensen & P. Madsen (eds.): Flexicurity and Beyond: Finding a new agenda for the European Social Model, Copenhagen: DJØF Publishing, pp. 335-364.

Madsen, P. K., Molina, O., Møller, J., Lozano, M. (2013), Labour Market Transitions of Young Workers in Nordic and Southern European Countries: The Role of Flexicurity. Transfer, 19(3), 325–343.

Maselli, I. (2010), Beyond Flexibility and Security: A composite indicator of flexicurity. CEPS Working Document No. 329/May 2010

MISSOC (2014), “Comparative Tables on Social Protection”, Situation on 1 July 2014, European Commission: www.missoc.org/MISSOC/INFORMATIONBASE/COMPARATIVETABLES/MISSOCDATABASE/comparativeTableSearch.jsp

Muffels, R., Wilthagen, T., Chung, H., Dekker, R. (2010), Towards a methodology to monitor and analyse flexicurity (FLC) and work-life balance (WLB) policies in the Member States of the EU. ReflecT Research Paper, 11/003.

O’Reilly, J., Eichhorst, W. Gabos, A., Hadjivassiliou, K., Lain, D., Lesckhe, J., McGuinness, S., Mýtna Kurakova, L., Nazio, T., Ortlieb, R., Russell, H. and Villa, P. (2015) ‘Five Characteristics of Youth Unemployment in Europe: Flexibility, Education, Migration, Family Legacies, and EU Policy’ SAGE Open Mar 2015, 5 (1) DOI: 10.1177/2158244015574962 http://sgo.sagepub.com/content/5/1/2158244015574962

van Oorschot, W. & Chung, H. (2015) "Feelings of dual-insecurity among European workers: A multi-level analysis" European Journal of Industrial Relations. 21(1): 23-37

Philips, K., Eamets, R. (2007) Approaches to Flexicurity: EU models. Luxembourg: European Foundation for the Improvement of Living and Working Conditions

Pissarides, C. A. (2009), The unemployment volatility puzzle: is wage stickiness the answer? Econometrica, Vol. 77, No. 5, 1339–1369.

Plantenga, J., Remery, C. and Rubery, J. (2007) “Gender mainstreaming of employment policies : A comparative review of thirty European countries” The co-ordinators’ synthesis report prepared for the Equality Unit, European Commission. July

Rubery, J., Smith, M., Fagan, C. and Grimshaw, D. (1998) Women and European Employment. London: Routledge

Rumler, F., Scharler, J. (2009), Labor Market Institutions and Macroeconomic Volatility in a Panel of OECD Countries, Working Paper Series, ECB Working Paper Series, No 1005/ February 2009.


Schmid, G. (2009) Transitional Labour Markets, from theory to policy application. Transitional Labour Markets and Flexicurity : Managing Social Risks over the Lifecourse. Documents de travail du Centre d'Economie de la Sorbonne 2009.75

Smith, M, Burchell, B., Fagan C. and O’Brien C (2008) “Job Quality in Europe” Industrial Relations Journal 39(6) 586-603

Smith M and Fagan C. (2008) “Analysis Note on Gender and Flexicurity” EU Directorate General Employment and Social affaires, Unit G1 “Equality between women and men. April 2008

Sperber, S. (2005), What are the ingredients of „good“ flexicurity arrangements? Some ideas for identifying factors that make for success. Trends in Social Cohesion No. 15: Reconciling labour flexibility with social cohesion – Facing the challenge, 191–210.

Tangian, A. (2007), European flexicurity: concepts, methodology and policies. Transfer: European Review of Labour and Research, 13: 551

Vielle, P., Walthery, P. (2003), Flexibility and social protection. Dublin, European Foundation for the Improvement of Living and Working Conditions.

Villa, P. (2007) “From “labour market flexibility” to “flexicurity” in the European Employment Strategy (EES)” Dipartimento di Economia, University of Trento Italy November 2007

van Vliet, O., Nijboer, H. (2012), Flexicurity in the European Union: flexibility for outsiders, security for insiders. Department of Economics Research Memorandum 2012.2, Leiden University.

Wilthagen, T. (2004), Balancing flexibility and security in European labour markets. Paper presented at the Conference on Recent Developments in European industrial relations, 7-8 October 2004.

Wilthagen, T. and Tros, F. (2004) The concept of ‘flexicurity’: a new approach to regulating employment and labour markets, TRANSFER: European Review of Labour and Research, Vol. 10 (2): 166-186.

Recent titles in this series

Available at: http://www.style-research.eu/publications/working-papers

WP3 Policy Performance STYLE Working Papers, WP3.1 Hadjivassiliou, K., L. Kirchner Sala, S. Speckesser (2015) Key Indicators and Drivers of Youth Unemployment STYLE Working Papers, WP3.2 Gonzalez Carreras, F., L. Kirchner Sala, S. Speckesser (2015) The Effectiveness of Policies to Combat Youth Unemployment STYLE Working Papers, WP3.3/DE Eichhorst, Wozny and Cox (2015) Policy Performance and Evaluation: Germany STYLE Working Papers, WP3.3/EE Eamets and Humal (2015) Policy Performance and Evaluation: Estonia STYLE Working Papers, WP3.3/NL Bekker, van de Meer, Muffels and Wilthagen (2015) Policy Performance and Evaluation: Netherlands STYLE Working Papers, WP3.3/PL Ślezak and Szopa (2015) Policy Performance and Evaluation: Poland STYLE Working Papers, WP3.3/ES González-Menéndez, Mato, Gutiérrez, Guillén, Cueto and Tejero (2015) Policy Performance and Evaluation: Spain STYLE Working Papers, WP3.3/SE Wadensjö (2015) Policy Performance and Evaluation: Sweden STYLE Working Papers, WP3.3/TR Gökşen, Yükseker, Kuz and Öker (2015) Policy Performance and Evaluation: Turkey STYLE Working Papers, WP3.3/UK Hadjivassiliou, Tassinari, Speckesser, Swift and Bertram (2015) Policy Performance and Evaluation: United Kingdom WP5 Mismatch: Skills and Education STYLE Working Papers, WP5.1 McGuinness, S., A. Bergin & A. Whelan (2015) A Comparative Time Series Analysis of Overeducation in Europe: Is there a common policy approach? STYLE Working Papers, WP5.3 Beblavý, M., Fabo, B., Mýtna Kureková, L. and Z. Žilinčíková (2015) Are student workers crowding out low-skilled youth?


STYLE Working Papers, WP5.4 McGuinness, S., A. Bergin and A. Whelan (2015) Recruitment Methods & Educational Provision effects on Graduate Over-Education and Over-Skilling. WP6 Mismatch: Migration STYLE Working Papers, WP6.3 Akgüç, M. and Beblavý, M. (2015) Re-emerging migration patterns: structures and policy lessons. WP7 Self-Employment and Business Start-Ups STYLE Working Papers, WP7.1 Sheehan, M. & McNamara, A. (2015) Business Start-Ups and Youth Self-Employment: A Policy Literature Overview. STYLE Working Papers, WP7.1/UK Hinks, R., Fohrbeck, A. & Meager, N. (2015) Business Start-Ups and Youth Self-Employment in the UK: A Policy Literature Review. STYLE Working Papers, WP7.1/DE Ortlieb, R. & Weiss, S. (2015) Business Start-Ups and Youth Self-Employment in Germany: A Policy Literature Review. STYLE Working Papers, WP7.1/IRL Sheehan, M. & Mc Namara, A. (2015) Business Start-Ups and Youth Self-Employment in Ireland: A Policy Literature Review. STYLE Working Papers, WP7.1/ES González Menéndez, M.C. & Cueto, B. (2015) Business Start-Ups and Youth Self-Employment in Spain: A Policy Literature Review. STYLE Working Papers, WP7.1/PL Pocztowski, A., Buchelt, B. & Pauli, U. (2015) Business Start-Ups and Youth Self-Employment in Poland: a Policy Literature Review. STYLE Working Papers, WP7.1/EE Masso, J. & Paes, K. (2015) Business Start-Ups and Youth Self-Employment in Estonia: A Policy Literature Review. WP8 Family Drivers STYLE Working Papers, WP8.1 Berloffa, G., M. Filandri, E. Matteazzi, T. Nazio, J. O’Reilly, P. Villa and C. Zuccotti, (2015) Work-poor and work-rich families: Influence on youth labour market outcomes.

34 Eamets, Beblavý, Bheemaiah Finn, Humal, Leschke, Maselli & Smith WP9 Attitudes and Values STYLE Working Papers, WP9.3 Hart, A., Stubbs, C., Plexousakis, S., Georgiadi, M., & Kourkoutas, E. (2015). Aspirations of vulnerable young people in foster care. WP10 Flexicurity STYLE Working Papers, WP10.1 Eamets, R., Beblavý, M., Bheemaiah, K., Finn, M., Humal, K., Leschke, J., Maselli, I. and Smith, M. (2015) Report Mapping Flexicurity Performance in the Face of the Crisis Key Indicators and Drivers of Youth Unemployment.

Forthcoming Publications in 2015 WP3 Policy Performance Policy Performance and Evaluation: Qualitative Country Case Studies (forthcoming) STYLE Working Papers, WP3.3 WP4 Policy Transfer Barriers to and triggers of innovation and knowledge transfer (forthcoming) STYLE Working Papers, WP4.1 WP7 Self-Employment and Business Start-Ups Mapping patterns of self-employment (forthcoming) STYLE Working Papers, WP7.2 WP8 Family Drivers Leaving and returning to the parental home during the economic crisis (forthcoming) STYLE Working Papers, WP8.3 WP9 Attitudes and Values Value system shared by young generations towards work and family (forthcoming) STYLE Working Papers, WP9.1 The impact of youth unemployment on social capital (forthcoming) STYLE Working Papers, WP9.2 WP10 Flexicurity From entry jobs to career employment (forthcoming) STYLE Working Papers, WP10.2 Flexicurity and Subjective Insecurity (forthcoming) STYLE Working Papers, WP10.3


Research Partners 1. University of Brighton – BBS CROME – United Kingdom

2. Institute for Employment Studies – United Kingdom

3. Institute for the Study of Labor – Germany

4. Centre for European Policy Studies – Belgium

5. TARKI Social Research Institute – Hungary

6. University of Trento – Italy

7. National University of Ireland Galway – Republic of Ireland

8. Democritus University of Thrace – Greece

9. University of Oxford – United Kingdom

10. Economic & Social Research Institute – Republic of Ireland

11. University of Salerno – Italy

12. University of Oviedo – Spain

13. University of Tartu – Estonia

14. Cracow University of Economics – Poland

15. Slovak Governance Institute – Slovakia

16. Metropolitan University Prague – Czech Republic

17. Grenoble Ecole de Management – France

18. University of Tilburg – Netherlands

19. University of Graz – Austria

20. Copenhagen Business School – Denmark

21. Norwegian Social Research – Norway

22. Swedish Instutute for Social Research – Sweden

23. Koç University Social Policy Centre – Turkey

24. University of Turin – Italy

25. EurActiv – Belgium

http://www.style-research.eu/research-organisations


Advisory Groups

Consortium Advisory Network

Business Europe www.businesseurope.eu

ETUI: European Trade Union Institute www.etui.org

European Youth Forum www.youthforum.org

European Foundation for the Improvement of Living and Working Conditions www.eurofound.europa.eu

ILO: International Labour Office www.ilo.org

OECD: Organisation for Economic Cooperation and Development www.oecd.org

OSE: Observatoire Sociale Européen www.ose.be

SOLIDAR: European network of NGOs working to advance social justice in Europe www.solidar.org EurActiv www.euractiv.com

European Commission, DG Employment, Social Affairs & Inclusion http://ec.europa.eu/social/main.jsp?langId=en&catId=1036

Local Advisory Boards including employers, unions, policy makers and non-government organisations

www.style-research.eu/project-advisors/local-advisory-boards/